Michael Wedin, Aristotle's Theory of Substance: The Categories and Metaphysics Zeta.   Oxford:  Clarendon Press, 2000.  Pp. x, 482.  ISBN 0-19-823855-X.  $55.00.  

A difficulty that has long confronted scholars of Aristotle's Categories and Metaphysics is the apparent contradiction between their respective accounts of primary substance. In the Categories, a primary substance (prôtê ousia) is the sort of thing that is in no way predicated of anything else: primary substances are the likes of Socrates, Callias, and other individual plants and animals (at least). When one reads the Metaphysics, however, it seems as if Aristotle has changed his mind. Aristotle's explicit statements about prôtê ousia suggest that it is not the individual as a whole but rather its form and essence (ti ên einai) that counts as a primary substance. Wedin's important new contribution to the literature devoted to this problem argues for a compatibilist account of the Categories and Metaphysics theories of substance. In the Metaphysics, the substance of the Categories (what Wedin calls a "c-substance") retains its "ontological" primacy as the subject of predications: the primacy of the form and essence is a new kind of "explanatory" primacy. Primacy in this sense belongs to the form because the latter explains the substantiality of c-substances. Hence, far from dropping out of the picture, c-substances serve as the explananda for the Metaphysics inquiry into substance. Wedin's conclusions will be of particular interest not only to scholars directly engaged with the question of the compatibility of the Categories and the Metaphysics, but also to those interested in the role of causal explanation in Aristotle's metaphysical project.

Wedin's argument occupies ten chapters. The first three are concerned specifically with the Categories, the fourth directly addresses the problem of the compatibility of the Categories with the Metaphysics, and the remaining chapters are aimed at constructing an account of Metaphysics Zeta that will show how it is compatible with the Categories.

Briefly, Wedin's account of the Categories is as follows. The Categories, Wedin argues, contains a "theory of underlying ontological configurations for standard categorical statements" (12), an account of the ontological implications of the various kinds of categorical statement. It should thus not be surprising that Wedin argues that this treatise is a more systematic work than is commonly acknowledged, beginning with his interpretation of the first chapter as introducing synonymy in order to use it as the foundation for the system of categories. Chapter 2 addresses the debate between Ackrill and Owen (and others) as to whether the things that are "present in" but not "said of" other things are individuals or universals, defending a refined version of Ackrill's interpretation aimed at preserving the position that such entities are at least nonrecurrent. This conclusion is important for chapter 3, which expounds the theory that (on Wedin's view) the Categories is intended to provide. Our ways of talking about the world commit us to a certain picture of what sort of things exist and of the relations in which they stand. An important component of this picture is that everything that is either said of or present in something else ultimately depends for its existence on something that is neither said of nor present in anything else, that is, on a primary substance. Aristotle's examples make it clear that the primary substances are individuals such Socrates, Secretariat, Madame Curie, and so forth.

Chapter 4 introduces the problem of the compatibility of the Categories account of primary substance with that of the Metaphysics by discussing the reasons why various interpreters have maintained their incompatibility. There is no doubt that the Categories and the Metaphysics differ significantly in their treatments of substance; the crucial thing for Wedin is that none of these differences amount to full-fledged incompatibility. Wedin's principal (though not his only) target here is the account of substance proposed in various places by Frede and Patzig, according to which the Metaphysics replaces the whole individual with its form as the basic subject for predications (129). Thus in the Metaphysics the form of a sensible substance becomes primary substance in the sense that the individual as a whole constituted primary substance in the Categories. Wedin criticizes the philosophical arguments that Frede and Patzig take to support this view as well as a more textual argument due to Christopher Shields.

Having addressed some prominent arguments for incompatibility in chapter 4, Wedin is in a position to begin his own compatibilist account of the Metaphysics in chapters 5-10. The exclusive object of Wedin's attention in these chapters is book Zeta (7), excluding chapters 7-9 and 12, which Wedin takes to be later additions and so outside the scope of the original "canonical" chapters of Zeta. The main object of Wedin's discussion of the early chapters of Zeta is to argue that the "what is substance?" question in Zeta should be read not as inquiring into what sorts of things are substances (i.e., as a "population" question), but as inquiry into the nature of substance and specifically the character of what Aristotle calls the "substance-of" things. Thus the discussion of the various candidates in Z3 aims to assess the suitability of each as the substance-of a c-substance, not their suitability as candidates for c-substantiality.

Wedin sees Z4 as introducing a new criterion for primacy in what he calls the "New Primacy" passage (the heart of which is at 1030a7-14), where Aristotle concludes that not everything has a definition but only those things that are primary (prôton), whose formulas are stated as not predicating one thing of another. Otherwise put, nothing that is not a genous eidos (1030a12) will be primary or count as an object of definition. Wedin devotes considerable argument to the position that genous eidos should be rendered not as "species of a genus" (a traditional rendering) but rather as "form of a genus", and so to the conclusion that Z4 explicitly introduces the explanatory primacy of the forms (and so the essences) of sensible substances. Part of Wedin's argument for this position construes the "impredicability" condition alluded to above as excluding predicating the differentia of the genus: Aristotle cannot be referring to a species of genus here in part because the impredicability criterion does not permit the genus-differentia predication to enter into a substance's form or essence (240).

Moving on to Z6, Wedin construes the identity of primary substance and essence demanded there as a condition of the explanatory power of essence in its capacity as the substance-of c-substances. If the "form of a genus" were not identical to its essence, it would suffer an unacceptable "dilution" of its explanatory power. The essence that was posited as explanatorily primary would turn out to be dependent on something else for what it is and so is no longer as credible a candidate for the role of substance-of. Another criterion for "substance-of-ness" is introduced in Z10-11: in order to be the substance-of a c-substance, form must be "purified" of matter. In contrast to some who allow the form of a substance to include high-level functional (or proximate) matter, Wedin argues primarily on textual grounds that Z10-11 demands that the form contain neither functional nor what Wedin calls "remnant" matter. The reason for this exclusion is found ultimately in Z17, where the form is shown to serve as the principle that transforms low-level matter into functional matter and so to make the matter "be" the c-substance whose matter it is (441-52). If the form is the cause of the transformation of low-level matter into functional matter, it cannot itself include functional matter on pain of making the explanandum part of the principle that explains it.

Apart from the conclusion that form must be pure, Wedin finds in Z10-11 the implication that form must be complex and that in its capacity as an explanatory entity it must somehow carry universality. As is well known, Z13 seems to cast doubt on both characterizations of form and serves as the primary evidence for those who maintain that primary substances must be particular forms. While avoiding the temptation to enter into the fray of the particular forms debate, Wedin seeks the preserve the possibility that form is universal by interpreting Z13 as endorsing what he calls "weak proscription." Unlike strong proscription, which maintains that no substance may be a universal, weak proscription maintains only that anything predicated universally of a c-substance (in practice, of a group of c-substances of the same species) cannot be the substance-of that c-substance (374). This interpretation is put into service later when Wedin discusses Z17. Since form is predicated not of a c-substance but of its matter, and, since the Z13 proscription applies only to something predicated of the c-substance as a whole, Z13 does not necessarily rule out the universality of form (426).

Whereas Z10-11 implies that the form will have parts and so must be complex, Z13 ends by concluding that if substances are not composed of universals--or more generally of other substances that exist in actuality--form will be simple and so not definable (1039a14-20). Aristotle promises to address the aporia created by this argument later (a22-3), which Wedin takes to refer to Aristotle's account of the parts of substances in Z16 (1040b5-15). Wedin takes the Z16 passage to advocate what he calls a "Dual Complexity": just as c-substances have parts which exist in the composite only potentially, so their forms will have parts corresponding to the parts of the composite, which exist in the form only potentially (cf. esp. b10-15).

Wedin brings his accounts of the various parts of Zeta together in his discussion of Z17. In addition to addressing questions peculiar to Z17, Wedin aims to show how the characteristics assigned to form and essence in the previous chapters contribute to its being an explanatory principle for c-substances. The explanandum is the substantiality of c-substances, that is, what Wedin calls the c-unity of matter that would otherwise be not a substance but a heap. In order to be the sort of principle that can explain the c-unity of a material substance, form must be pure of matter, and be something of a fundamentally different logical type from the matter it unifies. Furthermore, as the cause of something structurally complex, form too must have the sort of structural complexity that corresponds to the complexity of the substance it unifies. Form is primary substance in the sense that it possesses this explanatory power, which leaves c-substances in possession of the ontological primacy assigned to them in the Categories.

This contribution to the literature on Aristotle's account of substance is valuable for a number of reasons. Wedin's thesis is both important and intrinsically plausible, whether or not one ultimately agrees with it. No less valuable (and harder to appreciate from a description of the book's main theses in a short review) is the philosophical analysis of the Zeta texts presented in its support. In the more than 400 pages of the main text Wedin offers detailed and often compelling analyses of principal texts in the Categories and his "canonical" Zeta, and of the philosophical implications of the positions he finds in them. The book will thus likely hold considerable appeal not only for Aristotelian scholars but also for analytically trained metaphysicians with an interest in Aristotle. There is also to be found astute criticism of some leading alternative interpretations of these texts, especially those of Frede and Patzig.

The downside of Wedin's emphasis on the analysis of individual texts is a tendency to leave unaddressed certain important questions concerning the central books as a whole. For instance, Wedin simply appeals to the now-reputable status of the view that Z12 is an insertion to justify his leaving it out of his canonical books (343 n.1). This decision seems rather hasty given that Z12 bears directly on the question of the compositionality and definability of form, and given that Wedin is happy enough to use evidence from Z12 in support of his reading of genous eidos in Z4.¹ More seriously, Wedin never explains why he regards the whole of Eta as dispensable for his account of substantiality: once again, there are important passages in Eta that bear on issues he discusses (in addition to H6, over whose omission Wedin expresses regret, one might instance especially H3.1043b23-32 on definability and compositionality). The approach through the analysis of specific texts tends to downplay the importance of context, and Wedin's tendency to take every text as expressing Aristotle's considered view on an issue discourages questions about the dialectical character of parts of Zeta.² Thus in reading Z4 Wedin gives the impredicability condition a wide interpretation according to which it also forbids predication even within the category of substance, which together with an appeal to Z12 serves as evidence that the genus does not enter into the Z4 conception of essence. While I agree that this is ultimately Aristotle's position, I am doubtful that this is what Aristotle really means impredicability to entail in the context of Z4. To the extent that Zeta is dialectical, one might expect some statements to be first attempts at stating a position--subject to later refinement or clarification--and thus to be vague, ambiguous, or even contradictory of positions espoused elsewhere.

Criticisms aside, it is clear that this is an important book. Wedin has given us a careful, detailed, and insightful analysis of Zeta that will amply repay the careful study it demands.

1.   Wedin recognizes that the insertion of Z12 is not merely fortuitous, but maintains that if it is an insertion the canonical books should have a consistent story to tell without it. If one grants the stylistic evidence for regarding Z12 as an insertion, one may nevertheless wonder whether it was inserted at this point precisely because Aristotle thought it was necessary for the coherence of the central books' overall story.

2.   Aristotle's description of Z17 as sort of "new start" [1041a6-7] ought perhaps to bring to mind new starts elsewhere in the corpus made in the wake of dialectical discussion. See the opening lines of DA 2.1, NE 10.4, and EE 2.1.